Sensing of gas metal arc welding process characteristics for welding process control

Abstract

A method of sensing and controlling a gas metal arc welding process employs high frequency sampling of electrical signals from the welding circuit. The sampled signals are operated upon by predetermined processes to determine characteristics indicative of on-going weld quality and the future weld quality. The processes which operate upon sampled electrical signals determine shielding gas quality, the occurrence of short circuits and frequency thereof; pulse frequency and standard deviation thereof, contact tube wear and trends in the sampled electrical signals.

We claim: 
     
       1. A method of sensing and controlling a welding process carried out in a system including means for applying a weld using a contact tube, a power source to energize said means for applying a weld by forming a circuit therewith, means for sensing in-process welding characteristics and processing means for controlling said welding process in response to sensed in-process welding characteristics, said method comprising the steps of: (a) sampling an electrical signal reflecting in-process weld characteristics from said circuit;   (b) normalizing each sample with a calculated average signal in a sampling interval;   (c) determining power spectral density of said normalized sample signal;   (d) determining a contact tube wear sensitivity factor for said contact tube based upon an integrated value of said power spectral density;   (e) normalizing said contact tip wear sensitivity factor with a value of contact tube wear sensitivity for a new contact tube; and,   (f) stopping said weld process if said normalized contact tube wear sensitivity factor is beyond predetermined limitations.   
     
     
       2. The method of claim 1, wherein said electrical signals are sampled at a rate having a range of 8-250 Hz. 
     
     
       3. The method of claim 1, wherein said power spectral density is integrated from a frequency of 0 Hz to a predetermined cutoff frequency. 
     
     
       4. The method of claim 1, wherein current is sampled if said power source is voltage controlled. 
     
     
       5. The method of claim 1, wherein voltage is sampled if said power source is current controlled. 
     
     
       6. The method of claim 1, wherein said means for applying a weld carries out a gas metal arc welding process. 
     
     
       7. The method of claim 1, wherein said means for applying a weld carries out a flux core arc welding process. 
     
     
       8. A method of sensing and controlling a welding process to determine trends in said welding process, carried out in a system including means for applying a weld to material to be welded, a power source to energize said means for applying a weld by forming a circuit therewith, means for sensing in-process welding characteristics, and processing means for controlling said welding process in response to sensed in-process welding characteristics, said method comprising the steps of: (a) determining and storing an average value of signals from prior sampling intervals;   (b) determining a straight line based upon previously stored data for N sampling intervals using a statistical process;   (c) determining a confidence interval based upon said straight line;   (d) predicting a value of a signal sampled from said circuit for a future sampling interval and setting a confidence limit based upon said predicted value; and   (e) adjusting said welding process if a sampled electrical signal from a next sampling interval is outside said confidence limits.   
     
     
       9. The method of claim 8, herein said straight line is determined by a least square algorithm. 
     
     
       10. The method of claim 8, further comprising the step of comparing a slope of said straight line to predetermined limits to determine if sampled electrical signals are outside said confidence limits. 
     
     
       11. The method of claim 8, wherein the step (e) of adjusting the welding process comprises adjusting a wire feed device included with said means for applying a weld when sampled voltage is outside said confidence limits, and adjusting currents of said power source when sampled current is outside said confidence limits for a system using a constant current power source. 
     
     
       12. The method of claim 8, wherein the step (e) of adjusting the welding process comprises adjusting a wire feed device included with said means for applying a weld when sampled current is outside said confidence limits, and adjusting the power supply voltage when sampled voltage is outside said confidence limits for a system using a constant voltage power source. 
     
     
       13. The method of claim 8, wherein electrical signals are sampled at a rate having a range of 1 Hz-50 KHz. 
     
     
       14. The method of claim 8, wherein said means for applying a weld carries out a gas-metal-arc welding process. 
     
     
       15. The method of claim 8, wherein said means for applying a weld carries out a flux core arc welding process. 
     
     
       16. A method of sensing and controlling a welding process carried out in a system including means for applying a weld to material to be welded using a shielding gas, a power source to energize said means for applying a weld by forming a circuit therewith and generating pulses at selected frequencies, means for sensing in-process weld characteristics, and process means for controlling said weld process in processing means for controlling said weld process in response to sensed in-process weld characteristics indicative of shielding gas quality short circuit occurrence and frequency and pulse frequency and standard deviation thereof, said method comprising the steps of: (a) sampling an electrical signal reflecting in-process weld characteristics from said circuit;   (b) determining local, maximum and minimum values for said sampled electrical signal;   (c) determining a starting point for each pulse;   (d) determining an average pulse period based upon said start of each said pulse;   (e) determining a standard deviation of the average pulse period;   (f) determining an average pulse voltage for each pulse period;   (g) determining a mean of the average pulse voltage;   (h) determining a standard deviation of average pulse voltage based upon the mean of the average pulse voltage;   (i) calculating shielding gas condition based upon said average pulse voltage and said standard deviation of said average pulse voltage;   (j) comparing said gas condition to predetermined thresholds to determine if said weld process must be terminated;   (k) determining the occurrence of a short circuit based upon said local minimum value of said sampled electrical signal;   (l) determining an average short circuit frequency by calculating time intervals between short circuit occurrences; and,   (m) adjusting said power source in response to said average short circuit frequency.   
     
     
       17. The method of claim 16, wherein said electrical signals are sampled at a rate of 20-30 times per pulse. 
     
     
       18. The method of claim 16, wherein said power source is adjusted when said short circuit frequency is outside of predetermined parameters. 
     
     
       19. The method of claim 18, wherein said selected frequency of said power source is adjusted in response to said short circuit frequency, and current of said power source is adjusted if said power source is a constant current type, and voltage of said power source is adjusted if said power source is a constant voltage type. 
     
     
       20. The method of claim 16, wherein said means for applying a weld employs a gas metal arc welding process. 
     
     
       21. The method of claim 16, wherein said minimum signal point is found using the following process: (i) sweeping a window through an entire signal interval;   (ii) determining a parabola based upon signal data in said window and finding parabola coefficients;   (iii) determining if the parabola is upwardly concave;   (iv) determining a minimum parabola point;   (v) determining if an ordinate of the minimum point of said parabola lies within said window;   (vi) determining if an absicca of said minimum point of said parabola is a predetermined percentage less than end points of said parabola within said window; and   (vii) storing said minimum point in a local minimum array.   
     
     
       22. The method of claim 16, wherein said maximum signal value is determined using the following process: (i) sweeping a window through an entire signal interval;   (ii) determining a parabola based upon signal data in said window and finding parabola coefficients;   (iii) determining if the parabola is downwardly concave;   (iv) determining a maximum parabola point;   (v) determining if an ordinate of the maximum point of said parabola lies within said window;   (vi) determining if an absicca of said maximum point of said parabola is a predetermined percentage greater than end points of said parabola within said window; and   (vii) storing said maximum point in a local minimum array.   
     
     
       23. The method of claim 16, wherein the start of each pulse period is determined by the following process: (i) reading a logic signal from said power source;   (ii) selecting a data array index having a length less than that of a sample interval array; determining the presence of negative and positive crossings of a threshold;   (iii) determining if a negative crossing of said threshold occurs in less than a pulse interval;   (iv) incrementing said data array index where appropriate; and   (v) placing said data array index into an array of pulse period beginnings.   
     
     
       24. The method of claim 16 wherein the start of each pulse period is determined using the following method: (i) calculating an average of sampled electrical signals;   (ii) determining a threshold based upon a maximum signal value and said average signal;   (iii) selecting a data array index having a length less than that of a sample interval array; determining the presence of negative and positive crossings of said threshold;   (iv) determining if a negative crossing of said threshold occurs in less than a pulse interval;   (v) incrementing said data array index where appropriate; and   (vi) placing said data array index into an array of pulse period beginnings.   
     
     
       25. The method of claim 16, wherein said means for applying a weld employs a flux core arc welding process.